1
Gal Barak, MD; Shelley Kumar, MS, MSc; Geeta Singhal, MD, MEd Texas Children’s Hospital, Baylor College of Medicine Assessment of Clinical Learning Environment on a General Inpatient Pediatric Team On general inpatient pediatric teams, there are often a variety of learners, and the number of learners can vary greatly across different hospitals. Little is known about the number and composition of learners that best facilitates resident education. However, studies which assess team structure and size with regards to problem solving, often demonstrate negative correlations between team size and efficiency. As pediatric hospital medicine (PHM) fellowships have continued to develop, fellows are also playing a larger role in resident education. Multiple studies have demonstrated negative surgical residents’ perceptions towards fellows. Studies also demonstrated discrepancies between faculty and resident perception of surgical fellows. Little is known about non-surgical fellowships’ impact on residency training. Lim Y, Steinemann S, Berg B. Team Size Impact on Assessment of Teamwork in Simulation-based Trauma Team Training. Hawai’I Journal of Medicine & Public Health, 2014;73(11):358-361. Ofei-Dodoo S, Goerl K, Moser S. Exploring the Impact of Group Size on Medical Students’ Perceptions of Learning and Professional Development During Clinical Rotations. Kansas J Medicine, 2018;11(3):70-75. Plerhoples T, Greco R, Krummel T, Melcher M. Symbiotic or Parasitic? A Review of the Literature on the Impact of Fellowships on Surgical Residents. Annals of Surgery, 2012;256(6):904-908. To identify the optimal number and composition of learners on PHM teams To assess resident and faculty perceptions of the impact of a PHM fellow on resident learning National distribution to other comparable institutions to compare and contrast resident experiences and perspectives with TCH Qualitative data collection to identify factors driving these trends RESULTS Residents’ average satisfaction associated with the reported team size 0 1 2 3 4 5 0 2 4 6 8 10 12 Average Satisfaction Score Team Size r = - 0.35 Graph 6: Average satisfaction scores were calculated based on Likert scales for satisfaction in the following domains: receiving feedback, amount of time spent on teaching, amount of time spent with attending, level appropriate teaching, ability to ask questions, meeting the learning objectives, personal engagement during rounds, time for personal reflection, efficiency of rounds, and personally managing a sufficient number of patients. Satisfaction score can be predicted by 5.10 - 0.156 *Team Size. 0% 20% 40% 60% 80% 100% Current Optimal Current Optimal Percentage of responses Comparing Current and Optimal Team Sizes 0-2 3-4 5-6 7-8 >8 Resident Responses Attending Responses Graph 5: For residents as well as attendings, there was a significant difference in scores for current team size (M resident =7.0, M attending =6.3) and optimal team size (M resident =5.2, M attending =4.8) (p resident <.0001, p attendng =0.003). Male Male Female Female Other 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% Resident Faculty Percentage of responses Gender Graph 1: Study participants predominantly identified as female. PGY1 26% PGY2 36% PGY3 32% PGY4 6% Resident: Level of Training 0-3 11% 4-7 33% 8-10 11% 11+ 45% Faculty: Years out of residency Graph 2: Study participants represented a varied distribution of level of training (residents) and years out of residency (attendings). Figure 1: Average team size and composition as reported by residents and faculty, in comparison to the average optimal team size and comparison M M M S S I I I I U U F CURRENT OPTIMAL M M M S S I I I I U U CURRENT OPTIMAL F I I M S I U F = Medical Student = Sub-intern = Intern = Upper Level Resident = Fellow RESIDENT FACULTY Larger Team Smaller Team We hypothesize that there are more benefits to having a smaller rather than larger team Balancing Team Size Patient Care Education Autonomy Supervision TEAM Team structure plays a large role in implementing these components of resident training Key Aspects of Residency Training BACKGROUND RESULTS RESULTS OBJECTIVES METHODS CONCLUSIONS FUTURE DIRECTIONS REFERENCES 8% 15% 31% 23% 23% 13% 38% 50% 0% 20% 40% 60% 80% 100% Very negative Negative Neutral Positive Very positive Percentage of responses Fellow’s impact on resident learning experience Resident Attending p = 0.0232 Graph 3: There was a significant difference between resident and attending perspective on fellows’ impact. 35% 65% 93% 7% 0% 20% 40% 60% 80% 100% Yes No Percentage of responses Desire to work with fellow again Resident Attending p = 0.0006 Graph 4: There was a significant difference between resident and attending desire to work with a fellow again. Limited by sample size and reporting bias in an optional survey Cannot predict effect of intentionally decreasing team size, and barriers to implementation exist LIMITATIONS

Assessment of Clinical Learning Environment on a General ...Assessment of Clinical Learning Environment on a General Inpatient Pediatric Team On general inpatient pediatric teams,

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Assessment of Clinical Learning Environment on a General ...Assessment of Clinical Learning Environment on a General Inpatient Pediatric Team On general inpatient pediatric teams,

Gal Barak, MD; Shelley Kumar, MS, MSc; Geeta Singhal, MD, MEd

Texas Children’s Hospital, Baylor College of Medicine

Assessment of Clinical Learning Environment on a General Inpatient

Pediatric Team

On general inpatient pediatric teams, there are often a variety of

learners, and the number of learners can vary greatly across

different hospitals. Little is known about the number and

composition of learners that best facilitates resident education.

However, studies which assess team structure and size with

regards to problem solving, often demonstrate negative correlations

between team size and efficiency.

As pediatric hospital medicine (PHM) fellowships have continued to

develop, fellows are also playing a larger role in resident education.

Multiple studies have demonstrated negative surgical residents’

perceptions towards fellows. Studies also demonstrated

discrepancies between faculty and resident perception of surgical

fellows. Little is known about non-surgical fellowships’ impact on

residency training.

• Lim Y, Steinemann S, Berg B. Team Size Impact on Assessment of Teamwork in Simulation-based Trauma

Team Training. Hawai’I Journal of Medicine & Public Health, 2014;73(11):358-361.

• Ofei-Dodoo S, Goerl K, Moser S. Exploring the Impact of Group Size on Medical Students’ Perceptions of

Learning and Professional Development During Clinical Rotations. Kansas J Medicine, 2018;11(3):70-75.

• Plerhoples T, Greco R, Krummel T, Melcher M. Symbiotic or Parasitic? A Review of the Literature on the

Impact of Fellowships on Surgical Residents. Annals of Surgery, 2012;256(6):904-908.

• To identify the optimal number and composition of learners on

PHM teams

• To assess resident and faculty perceptions of the impact of a

PHM fellow on resident learning

• National distribution to other comparable institutions to compare

and contrast resident experiences and perspectives with TCH

• Qualitative data collection to identify factors driving these trends

RESULTS

Residents’ average satisfaction associated with the reported team size

0

1

2

3

4

5

0 2 4 6 8 10 12

Av

era

ge

Sati

sfa

cti

on

Sco

re

Team Size

r = - 0.35

Graph 6: Average satisfaction scores were calculated based on Likert scales for

satisfaction in the following domains: receiving feedback, amount of time spent on

teaching, amount of time spent with attending, level appropriate teaching, ability to ask

questions, meeting the learning objectives, personal engagement during rounds, time

for personal reflection, efficiency of rounds, and personally managing a sufficient

number of patients.

Satisfaction score can be predicted by 5.10 - 0.156 *Team Size.

0%

20%

40%

60%

80%

100%

Current Optimal Current Optimal

Pe

rce

nta

ge

of

res

po

ns

es

Comparing Current and Optimal Team Sizes

0-2

3-4

5-6

7-8

>8

Resident Responses Attending Responses

Graph 5: For residents as well as attendings, there was a significant difference in scores for current team size

(Mresident=7.0, Mattending=6.3) and optimal team size (Mresident=5.2, Mattending=4.8) (president<.0001, pattendng =0.003).

Male Male

Female Female

Other

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Resident FacultyP

erc

en

tag

e o

f re

sp

on

ses

Gender

Graph 1: Study participants predominantly

identified as female.

PGY1 26%

PGY2 36%

PGY3 32%

PGY4 6%

Resident: Level of Training

0-3 11%

4-7 33%

8-10 11%

11+ 45%

Faculty: Years out of residency

Graph 2: Study participants represented a varied distribution of

level of training (residents) and years out of residency (attendings).

Figure 1: Average team size and composition as reported by residents and faculty, in comparison to

the average optimal team size and comparison

M M M

S S

I I I I

U U

F

CURRENT OPTIMAL

M M M

S S

I I I I

U U

CURRENT OPTIMAL

F

I I

M

S

I

U

F

= Medical Student

= Sub-intern

= Intern

= Upper Level Resident

= Fellow RESIDENT FACULTY

Larger Team Smaller Team

We hypothesize that there are more benefits to

having a smaller rather than larger team

Balancing Team Size

Patient Care Education

Autonomy Supervision

TEAM

Team structure plays a large role in

implementing these components of

resident training

Key Aspects of Residency Training

BACKGROUND RESULTS RESULTS

OBJECTIVES

METHODS

CONCLUSIONS

FUTURE DIRECTIONS

REFERENCES

8%

15%

31%

23% 23%

13%

38%

50%

0%

20%

40%

60%

80%

100%

Verynegative

Negative Neutral Positive Verypositive

Perc

en

tag

e o

f re

sp

on

ses

Fellow’s impact on resident learning experience

Resident Attending

p = 0.0232

Graph 3: There was a significant difference

between resident and attending perspective on

fellows’ impact.

35%

65%

93%

7%

0%

20%

40%

60%

80%

100%

Yes No

Perc

en

tag

e o

f re

sp

on

ses

Desire to work with fellow again

Resident Attending

p = 0.0006

Graph 4: There was a significant difference

between resident and attending desire to work with

a fellow again.

• Limited by sample size and reporting bias in an optional survey

• Cannot predict effect of intentionally decreasing team size, and

barriers to implementation exist

LIMITATIONS